The 90 kDa heat shock proteins are proving to be extraordinary cancer chemotherapeutic targets as evidenced by the fact that more than 20 clinical trials are currently in progress. Unfortunately, all of these trials are based upon N-terminal inhibitors, primarily geldanamycin- derived, which exhibit serious formulation, scheduling and dosing difficulties as these compounds induce Hsp's at the same concentration they induce client protein degradation. Previous studies by Neckers and coworkers determined that Hsp90 contains a C-terminal ATP binding site that bound coumarin antibiotics competitively versus ATP. Like N-terminal inhibitors, inhibitors of the C-terminal binding domain also cause the degradation of Hsp90-dependent client proteins required for tumor cell growth and proliferation. A major drawback of the coumarin antibiotics is that they bind weakly to Hsp90 (IC50 approximately 700 micromolar); however, recent studies by our group have led to compounds >1000 fold more active than these natural products. Through extensive studies and elucidation of the C- terminal binding site, we have learned how to modulate Hsp90 in ways not previously realized. Thus, we have developed compounds that induce Hsp's at low concentrations that refold denatured proteins as a new method to treat various neurodegenerative diseases. In contrast, we have constructed molecules that inhibit Hsp90 without inducing Hsp's, and therefore provide a mechanism by which to bypass difficulties observed with N- terminal inhibitors in the clinic. In this application we propose to develop anticancer agents based on our recently elucidated C-terminal binding site in an effort to produce agents suitable for clinical evaluation. In addition, we plan to investigate the most active compounds in additional in vivo models of cancer